Re: Generic Modeling

From: Brian Smith <brian-l-smith_at_uiowa.edu>
Date: 31 Dec 2001 19:55:51 -0800
Message-ID: <60360d48.0112311955.2f5a353b_at_posting.google.com>


jraustin1_at_hotmail.com (James) wrote in message news:<a6e74506.0112311211.6223d0da_at_posting.google.com>...
> > ...generic modelling in relational databases...
>
> I have developed ATE systems and utility forecasting systems that used
> as few as 3 tables using Access and MS SQL Server. It could be
> described as things, properties and relationships tables. This
> provided the needed flexiblity but the performance was insufficient.

Was the performance bad no matter what you did? For database tuning, I don't know what MS SQL Server has. But Oracle has things like:

  • Index-Organized Tables (fast access to narrow rows by primary key)
  • paritioning (where you can split your data into partitions so that each partition holds the data that would have been in a seperate table using the "traditional" approach)
  • materialized views (which should allow you to maintain a relational view and a generic view of the data at the same time, automatically synchronized).
  • label security (you can control table access on a row-by-row basis).

I can't help thinking that by using these features one might be able to scale up to very large data sets without compromising on the design (i.e. have a generic model in physical tables and a relational model implemented as views of the generic model).

  ...
> Alternatively, XDb is a simple oodb that provides the type of
  ...

Have you personally used this system for any projects?

Thank you for the information.

  • Brian Smith
Received on Tue Jan 01 2002 - 04:55:51 CET

Original text of this message